Radiative and thermal effects of clouds

Clouds affect flows of energy from the Sun and the Earth in diverse ways and with sometimes opposite effects. The result of the reflective effect in terms of solar radiation (loss of energy) and the greenhouse effect in terms of terrestrial radiation (retention of energy) depends on a number of factors, including the size of the droplets, the density of the clouds, their thickness, altitude and temperature, among others.

It is believed that clouds’ net effect on the planet's surface is one of cooling. From differences in solar radiation reflected from cloudy and cloud-free skies, it has been deduced that clouds increase the planet's albedo (global reflectivity) by 15%, which means a loss of 50 W/m2. Scientists calculate that in compensation, clouds’ retention of outgoing infrared radiation creates a global increase, or greenhouse effect, of some 30 W/m2. The resulting effect is therefore negative: -20 W/m2.

Now, due to the immense variability in cloud cover and other unknown theoretical aspects regarding their microphysics, there is still a great deal of uncertainty surrounding this figure. The inclusion of cloud effects in climate models continues to be problematic and is subject to constant changes. In fact, an analysis of 19 different models shows that a dozen of them give very different figures, attributing to clouds a net cooling effect of more than -30W/m2 (Cess, 2005).

The radiative effect caused by clouds varies greatly from one region of the planet to another. The distribution of percentages of absorbed and reflected solar energy varies significantly according to the type of cloud, the latitude and the season of the year (Li, 1995). For example, in the tropical ocean regions of Western Africa and South America, frequently covered by low layers of stratocumulus, clouds can cause a net surface decrease of 100 W/m2. On the other hand, the high, fine cirrus clouds that sometimes cover tropical deserts tend to produce a net increase of up to 25 W/m2. In mid-level latitudes, intense depressions associated with cloud fronts tend to have a cooling effect, due to their very high albedo. In polar regions, on the contrary, cloud cover has a warming effect, since in addition to the greenhouse effect, the clouds have a lower albedo than the underlying surfaces, which are clear of clouds but covered in snow.

In the balance of radiation reaching the Earth’s surface, calculations based on satellite readings show that clouds produce a small temperature increase in the tropics, a very notable drop in temperature in mid-level latitudes, and a slight warming in high latitudes (Sohn, 1992).

It is also interesting to note that, although globally clouds have a cooling effect, their thermal effect on the surface is different during the day and at night. In general, clouds tend to have a cooling effect during the day and a warming effect at night, and therefore lead to a decrease in oscillation between daily maximum and nightly minimum temperatures. Clouds’ influence on daily thermal oscillation was corroborated on the three days following the destruction of the Twin Towers, during which planes were prohibited from flying over the United States. During those days, the diurnal temperature oscillation increased by more than 1° C. The most probable reason for this is the absence of the contrails (or vapour trails) left by planes as their hot exhaust gases facilitate condensation (Travis, 2002).

Figure 92. Airplane contrails over the southeast United States.

earthobservatory.nasa.gov

Types of clouds

According to Kirchoff’s law, bodies that absorb radiation are also emitters. Therefore, clouds also emit radiation, both downwards and upwards. According to the Stefan-Boltzmann constant, the total radiation emitted is proportional to the fourth power of the temperature. Because air temperature decreases with altitude, clouds are almost always colder than the Earth’s surface below them. Consequently, the infrared radiation emitted upwards by a blanket of clouds, allowing it to escape into space, is always smaller than the radiation emitted by the Earth’s surface that is retained in the atmosphere by this cloud cover. This is the key to clouds’ powerful greenhouse effect. Not only do they bounce part of the energy absorbed back down towards the surface, but also, the amount of energy they allow to escape upwards is always smaller than the quality of terrestrial infrared energy absorbed earlier. Not all clouds, however, behave in the same way.  Clouds with a higher temperature emit more radiation than cooler clouds; and as air temperature tends to decrease with altitude, lower clouds tend to emit more radiation than higher ones.

Figure 93. Radiative differences in clouds according to their height. Left:  low, thick, warm clouds reflect a good deal of sunlight (yellow arrows) and also emit infrared radiation (red arrows) upwards towards the exterior, thus cooling the Earth’s surface. Right: fine, high, cirrus clouds, made of small ice crystals, are transparent to solar radiation and their infrared emission towards space is low because their surface is very cold. They therefore have a warming effect on the surface.

Cirrus clouds

Scientists believe that, in general, high, thin cirrus clouds, which are very cold and formed by translucent ice crystals, allow a lot of incoming solar radiation to pass through (low albedo), but trap a good deal of the surface energy reaching them from beneath. This is because, due to their low temperature, they only emit and allow to escape into space a very small amount of energy. Thus, they add energy to the troposphere because they have a weaker albedo effect than greenhouse effect. But not all cirrus clouds are the same.

According to Ramanathan, who bases his theories on research into the effects of El Niño conditions in the Pacific, tropical clouds act as thermostats, curbing regional warming (Ramanathan, 1991). According to this theory, which has been rejected by other scientists (Mitchell, 1991), the surface temperature increase of the ocean can never exceed a certain limit because high temperatures provoke an increase in the convection and thickness of icy cirrus clouds; as a result, the clouds are no longer translucent and become highly reflective. These cirrus clouds, in anvil form, accumulate at the top of tropical cumulus clouds. Unlike translucent cirrus clouds, they now form an expansive cloud blanket that is opaque to solar radiation, covering a vast area and producing a cooling, or temperature control, effect on the ocean’s surface.

Laboratory experiments demonstrate that sulphate ammonium aerosols ((NH4)2SO4) produced by agricultural activities may play a significant role in the formation of cirrus clouds, since they turn water vapour into ice crystals. It should be noted, however, that these crystals, which can be fairly large, settle more quickly and dry up the air, thus reducing the greenhouse effect (Abbat, 2006).

Low stratus clouds

The low, white stratus clouds that often cover vast expanses over water reflect a huge amount of solar energy upwards and, in addition, retain very little terrestrial infrared radiation. Because their upper limits are located at low altitudes, the temperature of the surface emitting energy towards space is high, as is the upward-emitted energy value. These clouds have a cooling effect on the surface, which some believe to be around 15 W/m2.

Cumulus clouds

There is even more uncertainty regarding the effects of cumuliform clouds, as their absorption, reflection and energy emission percentages are unknown. This is due to the dependence of said values on physical characteristics, such as thickness, density and temperature at various levels.

Cloud cover and cosmic radiation

Galactic cosmic rays consist of particles with a very high energy (fundamentally protons) that originate in the supernovas of our galaxy, outside the Solar System. It is possible that the cosmic radiation that enters the Earth's atmosphere encourages, through various ionising processes, a greater concentration of condensation nuclei in the air and, consequently, the formation of more cloud cover (Carslaw, 2002). A recent experiment directed by Henrik Svensmark for the Danish National Space Centre succeeded in simulating this process (Svensmark, 2007).

Scientists know that an increase in the intensity of solar wind (also a flux of ionised particles but with less energy) diminishes the entrance of cosmic radiation into the entire Solar System, including the Earth’s atmosphere. By modifying the interplanetary magnetic field, solar winds act as a shield that rejects the entrance of intruding cosmic rays from other stars.

Greater solar activity means less cosmic radiation and less cloud cover; and less solar activity means more cosmic radiation and greater cloud cover. In this sense, scientists have confirmed that the total cloud cover over the Earth’s surface, measured by satellites since 1979, oscillates between 65% and 68%, and this variation seems to have coincided, in terms of low cloud cover (up to an altitude of 3 km), with variations in the incoming cosmic radiation reaching the Earth.

Solar winds influence geomagnetic activity on Earth. Over the last century, their intensity has taken an upward turn. This rise was accompanied by a decrease in galactic cosmic radiation and a smaller occurrence of low-lying clouds, which, according to this theory, would have caused an increase in the Earth's surface temperatures.

A Danish research team directed by Friis-Christensen (Friis-Christensen, 1991) has also found a positive correlation between the duration of solar cycles and 20th century temperatures. It has been observed that the mean temperature of the Earth’s surface rises during periods of short solar cycles, and falls when those cycles become longer. When cycles are short, the Sun is more active, and therefore brighter. There is more solar wind and less cloud cover. Thus, the Earth's surface warms up.

However, this theory regarding the relationship between cloud cover and cosmic rays has not yet been demonstrated satisfactorily, as satellite measurements of global cloud cover still span only a very short period. Furthermore, scientists have not taken into account the different effects that clouds have on temperature in accordance with their altitude. Nor are we able to explain why, despite the fact that the oscillation in the inflow of cosmic radiation is more intense in high latitudes, the correlation between cloud cover and radiation is more tenuous there than in mid-level or lower latitudes (Kernthaler, 1999).

A more recent theory relates solar variability with cloud cover not in relation to induced variations in cosmic ray intensity, but rather in connection with its influence on the formation of stratospheric ozone. Scientists know that in eleven-year solar cycles, changes in ultraviolet radiation (responsible for producing ozone) are relatively significant. A greater or smaller production of ozone ultimately influences stratospheric warming, and this warming indirectly affects tropospheric circulation. This last factor would alter cloud cover and explain the significant correlation between solar variability and the extent of cloud cover in regions like the United States (Udelhofen, 2001).

The evolution of global cloud cover

Clouds cover approximately 65% to 68% of the Earth's surface. This percentage varies according to temperature, humidity levels and the quantity of condensation nuclei present in the air. The small water droplets of which clouds are made always form in the presence of these nuclei. For this reason, all natural and man-made aerosols, as well as ionised particles linked to solar and cosmic radiation, have direct repercussions on changes in the Earth's cloud cover. For example, pollution from volatile particles increases low and mid-level cloud cover over large urban areas and a higher level of air traffic increases the layer of high, translucent cirrus clouds over some regions (Boucher, 1999; Stubenrauch, 2005).

How cloud cover will evolve, in which direction, and the effect that this evolution will have on the climate in the near future is the great unknown of computer models. Just about the only thing we know for certain is that, like a wild card hidden in a deck, cloud cover will significantly influence our climate.

In addition to the level of ‘dirtiness’ or the number of condensation nuclei in the air, cloud cover can also vary as the result of a change in the general atmospheric circulation. Cloud cover is greater in areas dominated by convergent and ascendant air currents, and is lower in areas with descendent, divergent ones. Therefore, changes in the general atmospheric circulation, which alter the extension and intensity of both low pressure (wind convergence) and high pressure (wind divergence) areas, can also influence cloud cover at a global scale.

From 1960 until the late 1980s, cloud cover seems to have increased above almost every continent. At a global scale, 86% of the world's stations recorded increases in cloud cover. The increase in cloud cover during this period is calculated at 10% in the United States and 5% in Europe (Henderson-Sellers, 1992). The phenomenon was termed ‘global dimming’. It seems that this dimming was particularly intense in built-up areas (Alpert, 2005). Scientists calculate that it was accompanied by more than a 6 W/m2 decrease in solar radiation, which would mean an increase of 2% in the Earth's albedo (Charlson, 2005).

The dimming which occurred between 1960 and 1987 should by rights have caused cooling, not the worldwide net temperature increase of 0.4° C registered in those years by the world’s network of surface thermometers. One way to resolve this contradiction is to assume that there was a decrease in evaporation on the continents, which, by causing the ground to lose less of its heat, would have increased both the temperature of the ground and that of surface air.

The greater cloud cover was also accompanied by a decrease in the diurnal/nocturnal temperature oscillation, as minimum temperatures in general increased more than maximum temperatures (Braganza, 2004).

Figure 94. From 1987 to 2001, worldwide cloud cover decreased by 4%. Since 2001 it has  again increased. Readings were taken by the ISCCP (International Satellite Cloud Climatology Project). During the period analysed, the mean percentage of the sky covered by cloud was 66.5%.

However, this trend seems to have changed over the last two decades. Since 1987, the evolution of cloud cover appears more complex (Wild, 2005). The ISCCP (International Satellite Cloud Climatology Project) suggests that from 1987-2001 cloud cover decreased by 4%, with one half of this drop being due to fewer low-lying clouds and the other due to fewer mid-level and high clouds. However, between 2001 and 2004, cover again increased, this time by 2% (Pallé, 2005 and 2006). A satellite study of the recent period from 1990-2003 fails to show any appreciable trend in Europe (Meerkótter, 2004). In China, observations over recent decades reveal diminishing cloud cover in almost all regions (Kaiser, 2000).

(prev)                              (back to contents)                              (next)